What’s new in AI: models, tools and techniques, verified
Top story
Amazon Web Services has introduced GPU partitioning on SageMaker Inference using NVIDIA Multi-Instance GPU technology to allow multiple models to share a single hardware-isolated accelerator.
Amazon SageMaker Inference now supports NVIDIA Multi-Instance GPU (MIG) partitioning, allowing users to divide a single physical GPU into independent partitions. This feature enables organizations to host multiple models on one accelerator with hardware-enforced isolation, improving utilization and reducing inference costs.
Source published
Nov 6, 18:01
→Detected by AI Newsroom
Nov 7, 09:59
→Verification completed
Nov 7, 10:00
→Published
Nov 7, 10:02
Amazon Web Services has announced the general availability of Web Answer on Amazon Bedrock AgentCore, offering developers a managed connector for agentic web search and citation synthesis. Alongside this launch, AWS outlined advanced enterprise governance architectures integrating Okta Cross App Access, AgentCore Gateway, and Cedar policies.
Amazon Web Services has published documentation detailing two governance and security features for Amazon Quick: Smart Approvals, which uses automated review to reduce routine prompts while preserving human oversight, and Deny by Default, which restricts AI capabilities upon release unless explicitly allowed.
Amazon Web Services has published an architectural guide for building a serverless Model Context Protocol server on Amazon ECS Fargate. The design aims to provide AI agents with metadata-driven access to enterprise data lakes without requiring custom per-dataset integration code.
Hugging Face has published details on Phonon TTS, an on-device text-to-speech model lineage featuring low latency and small parameter sizes. The organization also introduced a modular transformers approach intended to reduce code duplication when integrating new models.
Hugging Face has published details on Phonon TTS, an on-device text-to-speech model lineage featuring low latency and small parameter sizes. The organization also introduced a modular transformers approach intended to reduce code duplication when integrating new models.
Amazon Web Services has announced the general availability of Web Answer on Amazon Bedrock AgentCore, offering developers a managed connector for agentic web search and citation synthesis. Alongside this launch, AWS outlined advanced enterprise governance architectures integrating Okta Cross App Access, AgentCore Gateway, and Cedar policies.